Quality Control, Robust Design, and the Taguchi MethodKhosrow Dehnad In 1980, I received a grant from Aoyama-gakuin university to come to the United States to assist American Industry improve the quality of their products. In a small way this was to repay the help the US had given Japan after the war. In the summer of 1980, I visited the AT&T Bell Laboratories Quality Assurance Center, the organization that founded modern quality control. The result of my first summer at AT&T was an experiment with an orthogonal array design of size 18 (OA18) for optimization of an LSI fabrication process. As a measure of quality, the quantity "signal-ta-noise" ratio was to be optimized. Since then, this experi mental approach has been named "robust design" and has attracted the attention of both engineers and statisticians. My colleagues at Bell Laboratories have written several expository articles and a few theoretical papers on robust design from the viewpoint of statistics. Because so many people have asked for copies of these papers, it has been decided to publish them in a book form. This anthology is the result of these efforts. Despite the fact that quality engineering borrows some technical words from traditional design of experiments, the goals of quality engineering are different from those of statistics. For example, suppose there are two vendors. One vendor supplies products whose quality characteristic has a normal distribution with the mean on target (the desired value) and a certain standard deviation. |
Contents
MACROQUALITY WITH MICROMONEY | 23 |
USING DESIGN OF EXPERIMENTS | 31 |
OFFLINE QUALITY CONTROL PARAMETER DESIGN | 51 |
Copyright | |
12 other sections not shown
Other editions - View all
Common terms and phrases
accumulation analysis adjustment parameters ANOVA AT&T Bell Laboratories circuit design columns component control factors control parameters customer's design matrix design of experiments design parameters dispersion effects etching example expected loss factor levels Figure Genichi Taguchi Glossary equivalent identify integrated circuit Kackar kilohms linear loss function manufacturing imperfections manufacturing process mask dimension mean square offset minimize noise factors off-line quality control optimum orthogonal array output voltage parameter design experiment parameter design problems performance characteristic performance measure performance statistic performance variation PerMIA Phadke photoresist post-etch pre-etch line width process design product design product's performance quadratic loss quality control methods quality improvement quality philosophy reduce response robust design Section signal factor SN ratio solder sources of noise Spin Speed square offset voltage sum of squares Table Taguchi Method target value tion tolerance interval transistor transistor gain variables Viscosity wafer wave soldering window WNO WNO WNO